Several lines of evidence suggest that the serine/threonine kinase B-raf might be an attractive target for drug development in melanoma and numerous raf inhibitors (e.g. sorafenib, CHIR-265, SB590885) are now available for laboratory and clinical testing. In vitro exposure to sorafenib leads to the activation of caspases and the nuclear translocation of Apoptosis-lnducing Factor (AIF) in some, but not all, melanoma cell lines, suggesting that the susceptibility of melanoma cells to apoptosis induced by raf inhibitors may be genetically determined. A series of gene expression profiling experiments with 30 melanoma cell lines treated with sorafenib and CHIR-265 is proposed to address this hypothesis. The clinical activity of sorafenib is limited unless the drug is given with chemotherapeutic agents, in which case response rates as high as 40% have been observed in extensively pretreated patients. We have observed that sorafenib activates several compensatory survival pathways (i.e. JNK and p53) that undermine the lethality of the drug. The blockade of these pathways with pharmacologic inhibitors or siRNAs sensitizes melanoma cells to sorafenib-induced apoptosis. Sorafenib inhibits STATS in melanoma cells and this inhibition contributes to the apoptosis induced by the drug. We have also observed that the absence of DMA methyltransferase-1, an enzyme involved in the epigenetic silencing of RASSF1A and other genes, sensitizes melanoma cell lines to sorafenib. One of the goals of this application is to determine if these in vitro effects can be duplicated in tumor xenografts and tumor specimens from patients receiving a raf inhibitor. Such a result would provide a strong rationale for the combination of a raf inhibitor with a p53 antagonist (e.g. EL625), a STATS or DMA methyltransferase inhibitor (e.g. decitabine) in the treatment of melanoma, a concept we propose to test in the last years of this grant. Although the kinases inhibited by sorafenib and CHIR-265 overlap, there are differences that may confer superior antitumor activity on the latter agent. One of the goals of this application is the design and execution of a Phase I trial of CHIR-265, the extension phase of which will assess its clinical potential. The studies proposed in this application will assess the antitumor and antiangiogenic effects of this agent in melanoma patients and provide information as to how a raf inhibitor might be optimally combined with other targeted therapies to best exploit the potential of this novel class of drugs.